Donor-specific antibodies (DSA) continue to be a substantial barrier to renal transplantation. DSA are directed at MHC antigens (alloantibody), tissue antigens, or ABO blood group carbohydrates. These antibodies are thought to arise from germinal center (GC) and marginal zone (MZ) B cells, respectively. Over 18,000 pa- tients on the kidney transplant waiting list have significant alloantibody levels, and 25% of living kidney donors is excluded due to ABO incompatibility. Anti-ABO antibodies cause rejection of kidney transplants by binding to donor vascular endothelium, leading to macrovascular thrombosis and organ loss. To avoid rejection in sen- sitized or ABO incompatible recipients, pre-transplant immune desensitization (IMDS) is needed. IMDS proto- cols combine several therapies: (i) removal of circulating DSA by therapeutic plasma exchange (PLEX) (ii) pooled intravenous immunoglobulin (IVIG) therapy to alter the fractional catabolism of DSA, (iii) pharmacologic plasma cell depletion to suppress DSA synthesis, and (iv) B-cell depleting therapies to decrease donor-specific memory B cells. However, many patients fail IMDS when DSA levels cannot be reduced to a threshold safe for a kidney transplant or when side effects occur, especially bleeding and infection. No widely-accepted tools currently exist to guide the timing, dosing, and selection of IMDS agents. [The overall objective of this project is to develop a mathematical model of immune desensitization as a clinical tool to guide personalized therapy. The specific goals are: (1) to construct and validate a differential equation model of anti-ABO IgG kinetics and clinical therapies for immune desensitization, including exploration of alternative models; (2) to test the hypothesis that PLEX and IVIG therapies alter anti-ABO IgG B cell homeostasis, and to extend the model so it accounts for this feature, if needed; (3) to validate the ability of the model developed in Aims 1 and 2 to predict the response of individual patients to IMDS therapy and potential complications. We will further develop a compartmental model of IgG synthesis and catabolism kinetics which accommodates actual clinical protocols. Anti-ABO antibody levels and B cell phenotypes will be measured in patients undergoing PLEX, IVIG and plasma cell depleting therapies. This data will be used to develop and refine the model, and to estimate model parameters. Finally, we will evaluate how well the model predicts the success or failure of immune desensitization in an independent sample of patients undergoing standard immune desensitization for ABO incompatible kidney transplantation, which combines PLEX, IVIG, and plasma- cell depleting therapies. ]